- From: Lew Shannon <Lew.Shannon@TEAMInformatics.com>
- Date: Tue, 5 Apr 2005 10:37:38 -0400
- To: <public-xml-core-wg@w3.org>, "'Paul Grosso'" <pgrosso@arbortext.com>
>ACTION to Lew: Review the documents and report to >the WG by the end of March. My apologies to the group for not getting to this sooner. The customers keep getting in the way. I will plan to be on the call Wednesday. Here are my initial comments: At the 50K foot level, this is one of the most significant extensions to XML I think we have seen in some time. My congratulations to the authors in putting together a well thought-out and in-depth Use Case document and implementation plan. Although it is not clear as to how the various Use case scenarios map to the Binary Characterization Properties that are recommended. E.g, a significant increase in performance with binary encoded XML is being touted in the press. I don't see that as primary objective. A summary matrix (BCP, must haves, nice to have, etc.) with weighting and priorities would go a long way toward this end. The primary concern for out group will be the comprehensive set of recommended Binary Characterization Properties and how they propose to extend XML and existing processors as we know and love them. There are several much needed extensions including (compression, binary file embedding, and multimedia content management). However, several of the proposed extensions that will impact us and should be assessed are: - Alternative Serialization its impact on the info set. - Alternative Content type management and dependencies. - XML processors and Stack Integration - end-to-end processing model - Deltas vs Fragments - Adhoc Schema processing and data typing - Binary vs Text processing and encoded header fields Although BCP is a very comprehensive recommendation I see a lot of work on the part of the core group in accommodating it in terms of sorting the dependencies of the BCPs, how they play with the info set, and how to characterize/implement support in existing XML processors. Lew Principal, TEAM Informatics Inc.
Received on Tuesday, 5 April 2005 14:37:48 UTC